Funny dcarr, I was just thinking that I'd possibly be okay with this if a procedure such as the system redundancy typically used in parking garage structures was employed...
What I'm thinking is that the system would be required to meet ULS requirements in absence of all glass, possibly at a reduced load combination like 1.2D + 1.0L.
Using it for seismic only and having a full gravity system would seem reasonable, so long as we are talking about VERY robust glass systems.
Glass99: What type of glass panes would you be recommending for this kind of work? It makes me think of the polycarbonates I used in antiterrorism work. Lay up tempered on two faces with polycarb perhaps?
To all you with brittleness on your minds: would you design with stone or unreinforced masonry? Glass structures are way more engineered in respect to redundancy and progressive collapse than the typical brick wall that's perhaps holding up the roof over your head right now. -> its all about engineering around the brittleness.
In terms of impact, its not that hard to break the outer ply, but breaking subsequent plies takes a really substantial whack. Its more in the scale of multiple sledgehammer blows.
Nickel sulfide is not the thing it used to be. The float lines have really cleaned up their acts in respect to sulfur content. Also heat soak testing goes a long way. CELinOttawa: Cardinal had a widely cited study which spooked a bunch of folks about heat soaking in the US. People are getting over that now, primarily because heat soaked glass rarely gets NiS breakages. Its not 1995 any more!
KootK - yes I know John Kooymans from conferences mostly.
CELinOttawa: Bullet resistant glass has a polycarbonate core with glass on the outboard plies. I would not use this in a structural application typically, though you could. Its not as strong as all glass, plus the varying stiffness and coefficient of thermal expansion between polycarbonate and glass creates its own issues. Polycarbonate is actually much more brittle than people realize. Also bullet resistant glass is expensive. Simply tempered or heat strengthened glass, laminated with an SG interlayer is the right answer in practically all applications. The glass is one thing but a large part of the design is the connections and the structural system.
I think your analogy is a poor one. Unreinf masonry/stone often relies on its own weight and a thick cross section to keep stresses low. In that 1st pic it looked like the glass was the gravity load system as well. P-delta under some combinations and low out of plane buckling strength were my concerns. Clearly it can be done though.
dcarr: yes, its all about buckling where there's compression. Though buckling and P-Delta are easy to calculate. Glass is a linear elastic material just like steel. Pick a factor of safety and go for it! Its actually an easier calc than steel because there is no futzing with plastic hinges.
Yeah, my book on stone makes the same point. One of the assumptions of design is that reasonable proportions make stress irrelevant.
The part of glass design that really unmans me is discrete bolted connections. I think that connections is where the lack of ductility is the most dangerous. FEM on a bolt hole with the requirement to "know" maximum stresses is a scary prospect.
The greatest trick that bond stress ever pulled was convincing the world it didn't exist.
KootK - holes in glass are actually somewhat against nature. Done properly, they are gigantically strong - you break the bolt before you break the glass. Miss something about an imposed rotation or something and you are screwed! You may note that the Santa Fe house had no holes.
-> I like to tell clients - Glass is a fickle mistress. She wants the soft touch of the Spanish lover, not the rough hand of the Russian sailor.
I don't really do any work in glass but I am infinitely curious about using glass as a structural element and its design. What approach or design guide/manual do you typically follow? It seems like every design needs extensive FEA and testing outside of typical store front or curtain wall design. Do you work for a glass contractor or do you specialize in glass design, if you don't mind me asking. I'm just curious who does these type of designs.
I've read the book "Structural Use of Glass" by Matthias Haldimann which suggests a new approach to designing glass I think it was called the lifetime prediction model. I'm curious if that is used in practice?
RFreund: I am a consulting engineer, and have a small practice in NYC. I do work for contractors, owners, and architects. There are basically no code or books to follow, though there will be an ASTM for structural glass released at some point in the next few years, and there are some EuroNorms. I use first principles mostly and use finite elements for most designs. Personally, I hope there never is a code, because it will restrict use.
Kootk, i'm thinking of installing interior shear walls but i'm wondering since i've read a note that interior shear walls does not resist lateral force
Glass structures are very impressive. I have not designed anything in glass, but do know that the Institution of Structural Engineers publishes a guide book, "The structural use of glass in buildings".
Glass99 -> A few more questions (my apologies to the OP. I should probably start a new thread, I could be here all day or until you invite me to your office for bring your favorite eng-tiper to work day)
In regards to point glazing being very particular. Do recieve design properties to use in your FEM model from the point support manufacturers or is there somewhat of a standard? Is there much testing involved when you design these glass supported structures to verify these FEM models? I imagine there being many FEM models for just about each different piece of glass, do you guys have some sort of library of glass units that you modify? Sorry, all very fascinating to me. My father-in-law actually owns a small glass company, I've had distant dreams of constructing a glass bridge.
What role do you play on this project Lampi2k? Who's note is it that you've mentioned? I'd look into it. A note stating that shear walls don't resist lateral loads does't make much sense to me. What's the exact wording? Maybe there's a distinction between in plane and out of plane lateral loads.
The greatest trick that bond stress ever pulled was convincing the world it didn't exist.
Can you give us some shear values on that silicon product that you are stating makes your panels act like a shear wall? Pretty incredible photo by the way. Every architects dream. Also, I have never seen an infinitely strong allowable compressive stress in a material, except on my home planet of Krypton. Do you have some glass charts we could see with the allowable stress set to "infinity".
On a fibre by fibre basis, glass is stronger than most steel. It's only imperfection induced fracture mechanics stuff that reduces the strength drastically in tension.
The greatest trick that bond stress ever pulled was convincing the world it didn't exist.
Structural silicone has an allowable stress of 20psi for short term loads, which includes a factor of safety of approx 5 to 15 depending on the exact silicone.
There is technically a compressive strength of glass, but it is close enough to infinity. Its of the order of 1000ksi. Failure in compression structures is always a function of secondary tension stresses like poisson's ratio effects or buckling.
![[upsidedown]](/data/assets/smilies/upsidedown.gif "[upsidedown] [upsidedown]")
![[2thumbsup]](/data/assets/smilies/2thumbsup.gif "[2thumbsup] [2thumbsup]")